ABSTRACT Purpose: Modification of a donor cornea by gene therapy has potential to modulate irreversible rejection, the major cause of corneal graft failure. The sheep is a useful model for the human in this respect, as ovine endothelial cells are amitotic.The aim of the study was to investigate the ability of various non-viral and viral agents to transfer a reporter gene to ovine corneal endothelium. Methods: The non-viral agents Transfectin-10,Transfectin-20, Transfectin-50, SuperFect, Effectene and CLONfectin were used to deliver the reporter gene, Escherichia coli lacZ, to ovine corneal endothelium in vitro. A Herpes simplex virus-1 and an adenoviral vector each encoding E. coli lacZ were similarly tested. Infected corneas were organ-cultured for up to 7 days in vitro to allow transfection efficiency, duration of gene expression and toxicity attributable to each vector to be compared. Results: Scattered single or clusters of endothelial cells expressing the reporter gene were observed after trans- fection with CLONfectin,Transfectin-10,Transfectin-20 and Transfectin-50. SuperFect and Effectene were virtually in- effective. At best, the absolute number of infected cells per endothelial monolayer after 3 or 7 days of organ culture was estimated as < 0.01%. The Herpes simplex virus-1 vector also failed to transduce ovine corneal endothelium efficiently. In contrast, transfection rates of up to 70% of endothelial cells were observed with the adenoviral vector. Conclusion: Non-viral vectors and Herpes simplex virus-1 are unlikely to be suitable for gene therapy of corneal endothelium, because the efficiency of transfection is low compared with the rates achieved with adenoviral vectors. Key words: corneal endothelium, gene therapy, non-viral and viral vectors, reporter gene. INTRODUCTION The eye is an immune-privileged site; 1 however, irreversible immunological rejection remains the major cause of human corneal graft failure. 2 Rejection occurs despite the use of immunosuppression with topical glucocorticosteroids. 3,4 The major target of corneal graft rejection in humans is the corneal endothelium 5 and because these cells are essentially post-mitotic, damage cannot be repaired through cell divi- sion. 6 Gene therapy designed to reduce rejection-mediated damage holds some promise as a novel therapeutic strategy in corneal transplantation because the donor cornea can so easily be manipulated ex vivo, prior to transplantation. 7 Within Australia, Kaplan–Meier penetrating corneal graft survival is 59% at 10 years postoperatively and rejection accounts for approximately one-third to one-half of all corneal graft failures. 8 At a conservative estimate, any thera- peutic strategy capable of preventing irreversible rejection might have the potential to save approximately 50 grafts per year in this country. Furthermore, patients not currently offered a corneal graft because of their perceived high-risk status for rejection might be able to be offered surgery, were there to be a reasonable chance of a successful outcome. Among the decisions that need to be made in any exper- imental exploration of gene therapy for corneal transplanta- tion are choice of the animal model and choice of the vector for gene transfer. Orthotopic corneal transplantation in the sheep is a useful model for the purpose, in that the model is outbred. In the absence of topical immunosuppression, corneal allografts become spontaneously vascularized and undergo irreversible rejection at 3 weeks postoperatively in a manner that is clinically and histologically similar to human corneal graft rejection. 9 In particular, rejection is pre- ceded by a slight increase in anterior segment inflammation and both corneal epithelial and endothelial rejection lines are frequently observed. 9 An additional important consider- ation is that the ovine corneal endothelium is essentially amitotic. 10 In this respect it is like human corneal endothe- lium but unlike rodent corneal endothelium, which has a substantial replicative reserve. 11 Clinical and Experimental Ophthalmology (2001) 29, 316–322 Original Article Gene transfer to ovine corneal endothelium Sonja Klebe MB BS, 1 Pamela J Sykes PhD, 2 Douglas J Coster FRACO, 1 David C Bloom PhD 3 and Keryn A Williams PhD 1 Departments of 1 Ophthalmology and 2 Haematology and Genetic Pathology, Flinders University of South Australia, Bedford Park, South Australia, Australia, and 3 Department of Molecular Genetics and Microbiology, University of Florida College of Medicine, Gainesville, Florida, USA ■ Correspondence: Associate Professor KA Williams, Department of Ophthalmology, Flinders Medical Centre, Bedford Park, South Australia 5042, Australia. Email: keryn.williams@flinders.edu.au